From Mesh Generation to Scientific Visualization: An End-to-End Approach to Parallel Supercomputing

TLDR

Parallel supercomputing has traditionally focused on the solver kernel, while front‑end and back‑end tasks such as problem definition and output interpretation have received little attention to scalability, a gap that worsens as simulations scale to petascale. The authors propose an end‑to‑end parallel approach that tightly couples meshing, partitioning, solver, and visualization, eliminating intermediate I/O. They implement this approach using an octree‑based finite element model for earthquake ground motion, coupling all stages in parallel. On 2048 processors, the end‑to‑end method overcomes traditional scalability bottlenecks.

Abstract

Parallel supercomputing has traditionally focused on the inner kernel of scientific simulations: the solver. The front and back ends of the simulation pipeline problem description and interpretation of the output have taken a back seat to the solver when it comes to attention paid to scalability and performance, and are often relegated to offline, sequential computation. As the largest simulations move beyond the realm of the terascale and into the petascale, this decomposition in tasks and platforms becomes increasingly untenable. We propose an end-to-end approach in which all simulation components-meshing, partitioning, solver, and visualization-are tightly coupled and execute in parallel with shared data structures and no intermediate I/O. We present our implementation of this new approach in the context of octree-based finite element simulation of earthquake ground motion. Performance evaluation on up to 2048 processors demonstrates the ability of the end-toend approach to overcome the scalability bottlenecks of the traditional approach.